JP2008082871A - Obstacle detector for vehicle - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an obstacle detector which more accurately predicts the position of a target. <P>SOLUTION: The obstacle detector comprises: an obstacle detecting means 11 for detecting the target in front of one's own vehicle; and an operation apparatus controlling means 12 for receiving information of a detection target and controlling an operation apparatus of the own vehicle. The obstacle detecting means 11 comprises: a target data extracting section 111 for extracting target data including the distance to the target, relative velocity to that of the target, and the direction of target every predetermined sampling time; a target position predicting section 112 for predicting the target position at the next sampling time based on target data; and a target identifying section 113 for collating the position of the detection target with the position of the prediction target and identifying the same target. The target data extracting section 111 further determines a lateral velocity component of the target, and the target position predicting section 112 predicts the target position using the relative velocity and the lateral velocity component. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to an obstacle detection device for a vehicle that detects an obstacle ahead of the host vehicle, and more specifically, for a vehicle that predicts the position of a target using a lateral movement component of the target. The present invention relates to an obstacle detection device.

  In recent years, various obstacle detection devices that detect obstacles ahead of the host vehicle have been proposed for adaptive cruise control (ACC), collision prediction, and the like. In these obstacle detection devices, for example, a millimeter wave radar wave is emitted toward the front of the vehicle, and an obstacle (target) is detected by receiving the reflected wave. Further, the following Patent Document 1 discloses a technique for improving the reliability of obstacle detection by appropriately estimating the relative relationship between the obstacle and the vehicle when the obstacle is no longer detected by the radar device. It is disclosed.

Japanese Patent Laid-Open No. 7-215147

  In the millimeter wave radar apparatus, the position at the next sampling is predicted for each detected target. In the position prediction, the movement amount obtained by multiplying the target speed by a predetermined time is displaced from the current position of the target.

  By the way, in the radar apparatus, the speed of the target is detected as a relative speed between the host vehicle and the target on a straight line passing between the host vehicle and the target. For this reason, the predicted position of the target after a predetermined time may differ from the actual target detection position after the predetermined time. In particular, when another vehicle crosses in front of the host vehicle, the predicted position may be greatly different from the actual position.

  As a result, the same target may be misrecognized as another target at the next sampling, and the continuity of target detection may be lost. In addition, a collision may be predicted based on an incorrect predicted position of the target. In this case, the pretensioner mechanism or the like malfunctions even though the possibility of a collision is low in reality.

  Therefore, an object of the present invention is to provide an obstacle detection device that can predict the position of a target more accurately.

  In order to achieve the above object, an obstacle detection device for a vehicle according to the present invention detects an object in front of the host vehicle, and receives information on the detected target from the obstacle detection means. Operating device control means for controlling the operating equipment of the host vehicle, and the obstacle detecting means is a relative between the host vehicle and the target on a straight line passing through the host vehicle and the target at every predetermined sampling time. A target data extraction unit that extracts target data including speed, and a target position prediction unit that predicts the position of the target at the next sampling based on the target data. The unit further calculates the velocity component in the lateral direction with respect to the direction along the straight line passing through the vehicle and the target of the target, and the target position prediction unit uses the velocity component in the lateral direction along with the relative velocity. And predicting the position of the target.

  According to the obstacle detection device for a vehicle of the present invention configured as described above, the position of the target is predicted using the speed component in the horizontal direction together with the relative speed. Thereby, the position of the target can be predicted more accurately.

  Preferably, in the present invention, the target data extraction unit calculates a lateral velocity component of the target from an angle difference between the direction of the target at the time of the previous sampling and the direction of the target at the time of the current sampling. Ask. Thereby, the velocity component in the lateral direction can be easily obtained.

  In the present invention, it is preferable that the target data extraction unit obtains a lateral velocity component of the target from an average value of angle differences of the direction of the target at the time of sampling three times or more. As a result, a more accurate lateral velocity component can be easily obtained.

Preferably, in the present invention, the operating device control unit includes an interpolation prediction unit that interpolates and predicts the position of the target at a time interval shorter than the sampling interval of the obstacle detection unit, and the interpolation prediction unit includes the vehicle and the object. Using the speed component in the direction transverse to the direction along the straight line passing through the vehicle and the target along with the relative speed between the vehicle and the target on the straight line passing through the target, Is predicted by interpolation.
In this way, in the interpolation prediction of the target, if the velocity component in the horizontal direction is used together with the relative velocity, the position of the target is interpolated and predicted more accurately.

  In the present invention, preferably, the operating device is a brake operating means for operating a brake. By predicting the position of the target more accurately, the collision prediction accuracy is improved and the occurrence of brake malfunction is suppressed.

  In the present invention, preferably, the operating device is a pretensioner mechanism that winds up the seat belt so as to restrain the vehicle occupant with a predetermined tension. By predicting the position of the target more accurately, the collision prediction accuracy is improved, and the occurrence of malfunction of the pretensioner mechanism is suppressed.

  Thus, according to the vehicle obstacle detection device of the present invention, the position of the target can be predicted more accurately.

Embodiments of an obstacle detection device for a vehicle according to the present invention will be described below with reference to the accompanying drawings.
First, with reference to the block diagram of FIG. 1, the structure of the vehicle obstacle detection and transmission of the embodiment will be described. As shown in FIG. 1, the vehicle obstacle detection device 1 according to the embodiment includes an obstacle detection unit 11 that detects a target ahead of the host vehicle, and information on a target detected from the obstacle detection unit 11. And operating device control means 12 for controlling the operating device of the host vehicle.

  In FIG. 1, a brake operating means 2 and a pretensioner mechanism 3 are shown as operating devices. The brake operating means 2 operates an automatic brake at the time of collision prediction. The pretensioner mechanism 3 winds up the seat belt so as to restrain the vehicle occupant with a predetermined tension at the time of collision prediction.

  Further, the obstacle detection means 11 is composed of a millimeter wave radar device, and includes a target data extraction unit 111, a target position prediction unit 112, and a target marker separate unit 113. Further, the operating device control means 12 is constituted by an ECU (electric control unit) 1.

  In the target data extraction unit 111, for each predetermined sampling time, a distance L from the host vehicle to the target, a relative speed Vs between the host vehicle and the target on a straight line passing between the host vehicle and the target, Target data including the direction θ of the target viewed from the host vehicle is extracted. The sampling time interval corresponds to the radar sweep time and is, for example, 100 milliseconds (ms). The distance L to the target is determined by the time difference from the transmission to the reception of the radar wave. Further, the relative speed is obtained by a Doppler shift of the frequency of the received wave with respect to the frequency of the transmitted wave. Then, the current position of the target is obtained from the distance L to the target and the direction θ of the target.

  The target position prediction unit 112 predicts the position of the target at the next sampling based on the target data. The target position after the predetermined time is a position where the target is moved by the calculated movement amount from the current position of the target. The amount of movement is obtained as the product of the speed of the target and time.

  The target marking distinction unit 113 compares the detected target position with the predicted target position to identify the same target. That is, a target detected near the predicted position at the next sampling is recognized as the same target as the target whose position is predicted. Specifically, management is performed with the same identification number.

  Conventionally, in the target position prediction, only the relative speed Vs between the host vehicle and the target on a straight line passing between the host vehicle and the target is used as the target speed. For this reason, conventionally, an error has occurred in the predicted position of the target. In particular, when an oncoming vehicle is detected by a curve or when another vehicle crosses the front of the host vehicle, the actual position and the predicted position of the other vehicle after a predetermined time may differ greatly.

Here, with reference to FIG. 2, the target position prediction will be briefly described.
FIG. 2 shows a state where another vehicle J (0) crosses from the right to the left in front of the host vehicle S on the three-way road. The position of the target detected at the time of this sampling is shown as other vehicle J (0). For example, “ID1” is assigned to the detected other vehicle J (0) as the identification number of the target. Note that the current position of the other vehicle J (0) is obtained by, for example, the distance from the host vehicle S and the direction θ.

  In predicting the position of the other vehicle J (0) at the next sampling after 100 ms, conventionally, only the relative speed Vs between the host vehicle and the target on the straight line D passing through the host vehicle and the target is used. I was using it. For this reason, in the conventional prediction method, the predicted position of the other vehicle after 100 ms is the position indicated by P (1) in FIG.

  On the other hand, the actual position of the other vehicle after 100 ms is the position indicated by J (1) in FIG. Thus, since the actual position of the other vehicle at the next sampling is significantly different from the predicted position, the other vehicle J (1) detected at the next sampling is the other vehicle J (0) detected at the current sampling. It is recognized as a different target. As a result, for example, “ID2” is given to the other vehicle J (1) as a new identification number.

  Further, at the time of the next sampling, naturally, no other vehicle is detected in the vicinity of the predicted position P (1). However, it is common that the target once detected is not detected at the next sampling due to noise or the like. For this reason, the radar apparatus extrapolates the predicted position of the target with the identification number “ID1” and obtains it a plurality of times even if the target corresponding to the vicinity of the predicted position P (1) is not detected. As a result, the predicted position P (2) of the target “ID1” at the next sampling 200 ms after the current sampling, the predicted position P (3) after 300 ms, and the predicted position P (4 after 400 ms) ) Are calculated one after another.

  As a result, in the conventional position prediction method, the predicted position by extrapolation of another vehicle rapidly approaches the host vehicle. As a result, in reality, a collision with another vehicle may be predicted by mistake even though the other vehicle is not approaching the host vehicle. In that case, the brake is automatically actuated to avoid danger, the seat belt pretensioner mechanism is actuated to protect the occupant, the seat belt is taken up suddenly, or the airbag is deployed. There are things to do. In order to avoid such malfunction of the in-vehicle device, it is necessary to accurately predict the position of the other vehicle.

  Therefore, in the present invention, as shown in FIG. 3, the target data extraction unit 111 further obtains a velocity component Vt in the lateral direction with respect to the direction along the straight line passing through the subject vehicle and the target. . The lateral velocity component Vt of the other vehicle J (0) is detected at the time of sampling this time, for example, the direction θ (-1) of the other vehicle J (-1) detected at the previous sampling by the radar device. It may be obtained from the angle difference Δθ from the direction θ (0) of the other vehicle J (0). Moreover, the velocity component Vt in the horizontal direction of the target may be obtained from an average value of the angle difference between the directions of the target at the time of sampling three times or more. Further, the lateral speed component Vt may be obtained from the product of the angle difference Δθ and the distance L to the other vehicle J (0).

  Further, the target position prediction unit 112 predicts the position of the target by using the horizontal speed component Vt together with the relative speed Vs. Specifically, as shown in FIG. 3, the speed vector V1 of the other vehicle, that is, the sum of the speed vector of the relative speed Vs directed in the direction of the host vehicle and the speed vector of the speed component Vt in the lateral direction, The magnitude and direction of the speed is determined. Then, the position of the other vehicle after 100 ms is predicted from the current position of the other vehicle J (0) to a position moved by the amount of movement of the product of the speed vector V and the sampling time interval. As a result, the predicted position of the other vehicle J (0) is accurately predicted in the vicinity of the actual position of the other vehicle J (1) after 100 ms indicated by J (1) in FIG. As shown in FIG. 3, the predicted position P (1) using only the relative speed Vs and the predicted position J (1) also using the lateral speed component Vt are greatly separated.

  As a result of accurate position prediction in the present embodiment, the other vehicle J (1) is actually detected in the vicinity of the predicted position of the other vehicle J (0) at the next sampling. For this reason, the other vehicle J (1) detected at the next sampling is recognized as the same target as the other vehicle J (0) detected at the current sampling. As a result, the same identification number “ID1” as the other vehicle J (0) is assigned to the detected other vehicle J (1). Thereby, the detected target is appropriately managed.

  In addition, the accurate position prediction of the target avoids erroneous extrapolation of position prediction. As a result, erroneous collision prediction is avoided and the occurrence of malfunction of the in-vehicle device is prevented. In addition, an actual target is not detected in the vicinity of the predicted position after extrapolation a predetermined number of times, and losing the target (lost) is reduced.

  As shown in FIG. 1, the operating device control unit 12 of this embodiment includes an interpolation prediction unit 121 that interpolates and predicts the position of a target at a time interval shorter than the sampling interval of the obstacle detection unit 11. . The interpolating prediction unit 121, along with the relative speed Vs between the host vehicle and the target on the straight line passing through the host vehicle and the target, is transverse to the direction along the straight line passing through the host vehicle and the target. The position of the target is interpolated using the velocity component Vt in the direction.

  When the sampling time interval of the radar device of the obstacle detection means 11 is 100 ms, for example, a vehicle traveling at 72 km / h travels 2 meters during the sampling time interval. For this reason, it is preferable to perform collision prediction at shorter time intervals in order to operate vehicle travel control such as automatic braking and occupant protection devices such as airbags and pretensioner mechanisms. Therefore, the operating device control means 12 interpolates and predicts the position of the target at 20 ms intervals shorter than the sampling time interval 100 ms of the radar device.

  FIG. 4 schematically shows the predicted positions when interpolation prediction is performed using only the detected relative velocity Vs of the target and when interpolation prediction is performed using the relative velocity Vs and the lateral velocity component Vt. Shown in In the example shown in FIG. 4, interpolation prediction is performed four times at 20 ms intervals for the other vehicle J (0) detected during the current sampling until the next sampling.

  In FIG. 4, the predicted positions after 20 ms, 40 ms, 60 ms, and 80 ms when interpolation prediction is performed using only the relative velocity Vs are respectively P (01), P (02), P (03), and P ( 04). On the other hand, the predicted positions after 20 ms, 40 ms, 60 ms, and 80 ms when the relative speed Vs and the lateral speed component Vt are used are J (01), J (02), and J (03, respectively. ) And J (04). As shown in FIG. 4, as the prediction time becomes longer, the predicted position using only the relative speed Vs and the predicted position using the relative speed Vs and the lateral speed component Vt are separated from each other. At the time of the next sampling after 100 ms, the other vehicle is detected in the vicinity of the predicted position indicated by J (1), not P (1).

  With reference to the flowchart of FIG. 5, the processing contents of the interpolation prediction using the lateral speed component in the operating device control means 12 will be described. As shown in FIG. 5, the interpolation prediction unit first acquires detected target data from the obstacle detection means 11 (step S1). For example, the data of the other vehicle J (0) shown in FIG. 4 is acquired.

  Next, the target data is processed (step S2). Here, the lateral speed component of the other vehicle J (0), the relative speed, and the distance to the other vehicle J (0) are calculated. The lateral component may be obtained from, for example, the detection position of the other vehicle at the previous sampling and the detection position of the other vehicle at the current sampling, as performed in the obstacle detection means 11 described above.

  Next, the position of another vehicle after 20 ms is predicted by interpolation (step S3). The amount of movement in the horizontal direction can be obtained as a product of the velocity component Vt in the horizontal direction and a time of 20 ms. Further, the amount of movement in the vertical direction along the straight line connecting the other vehicle and the host vehicle can be obtained by the product of the relative speed Vs and the time 20 ms. Then, the position moved by the amount of movement in the horizontal direction and the vertical direction from the position of the target detected at the time of sampling this time becomes the position speed position after 20 ms. Similarly, positions after 40 ms, 60 ms and 80 ms are predicted.

  In this way, accurate prediction is possible by performing interpolation prediction using the velocity component Vt in the horizontal direction. Thereby, the collision prediction accuracy is improved, the occurrence of malfunction of the brake is suppressed, and the occurrence of malfunction of the pretensioner mechanism is suppressed.

  In each embodiment mentioned above, although the example which constituted the present invention on specific conditions was explained, the present invention can perform various change and combination, and is not limited to this.

It is a block diagram explaining the structure of the obstacle detection apparatus of the vehicle of embodiment of this invention. It is explanatory drawing of the position prediction of the target in an obstruction detection means. It is explanatory drawing of the position prediction of the target in an obstruction detection means. It is explanatory drawing of the interpolation prediction of the target position in an action | operation apparatus control means. It is a flowchart explaining the process in the interpolation estimation part of an action | operation apparatus control means.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Obstacle detection apparatus 2 Brake operation means 3 Pretensioner mechanism 11 Obstacle detection means 12 Actuator control means 111 Target data extraction part 112 Target position prediction part 113 Object label separate part 121 Interpolation prediction part

Claims (6)

Obstacle detection means for detecting a target ahead of the host vehicle;
Receiving the information of the detected target from the obstacle detection means, and operating device control means for controlling the operating device of the host vehicle, and
The obstacle detection means includes
A target data extraction unit for extracting target data including a relative speed between the host vehicle and the target on a straight line passing through the host vehicle and the target at a predetermined sampling time;
A target position prediction unit that predicts the position of the target at the next sampling based on the target data;
The target data extraction unit further calculates a velocity component in a lateral direction with respect to a direction along a straight line passing through the host vehicle and the target of the target,
The obstacle detection device for a vehicle, wherein the target position prediction unit predicts the position of the target using the lateral speed component together with the relative speed. The target data extraction unit obtains a lateral velocity component of the target from an angle difference between the direction of the target at the time of the previous sampling and the direction of the target at the time of the current sampling. Item 1. A vehicle obstacle detection device according to Item 1. 3. The vehicle according to claim 1, wherein the target data extraction unit obtains a speed component in a lateral direction of the target from an average value of an angle difference of the direction of the target at the time of sampling three times or more. Obstacle detection device. The operating device control unit includes an interpolation prediction unit that performs interpolation prediction of a target position at a time interval shorter than a sampling interval of the obstacle detection unit,
The interpolating prediction unit is configured so that the relative direction between the host vehicle and the target on a straight line passing through the host vehicle and the target, as well as the direction of the target along the straight line passing through the host vehicle and the target. The obstacle detection apparatus according to any one of claims 1 to 3, wherein the position of the target is interpolated using the velocity component. The vehicle obstacle detection device according to any one of claims 1 to 4, wherein the operating device is a brake operating unit that operates a brake. The vehicle obstacle detection device according to any one of claims 1 to 5, wherein the operating device is a pretensioner mechanism that winds up a seat belt so as to restrain a vehicle occupant with a predetermined tension. .

Images